How to Solve a Combinatorial Optimization Problem by using ACO
Given are the few guidelines suggested in order to explain the combinatorial optimization problem by using ant colony algorithm.
(a) Show the problem in the form of sets of components and transitions or by means of a weighted graph such is travelled by the ants to build solutions.
(b) Appropriately explain the meaning of the pheromone trails. It is a crucial step in the implementation of an Ant Colony Optimization algorithm.
(c) Appropriately explain the heuristic preference to all decisions such an ant has to take while constructing that is explain the heuristic information is crucial for good performance if local search algorithms are accessible or cannot be applied.
(d) Implement an efficient local search algorithm, if possible for the problem under taken, since the results of many Ant Colony Optimization applications to NP- hard combinatorial optimization problems illustrate the best performance is achieved when computing Ant Colony Optimization along with local optimizers.
(e) Select an exact Ant Colony Optimization algorithm and apply this to the problem being solved, taking the previous aspects into account.
(f) Tune the parameters of the Ant Colony Optimization algorithm. A good beginning point for parameter tuning is to employ parameter settings such were found to be good while applying the Ant Colony Optimization algorithm to similar problems or to a variety of other problems. An option to time-consuming personal involvement in the tuning task is to employ automatic process for parameter tuning.
Ant Colony Optimization: IMPLEMENTATION TO THE PROCESS PLANNING PROBLEM
So as to implement Ant Colony Optimization, a procedure planning problem has been shown in a graphical form; where, all nodes act as the machining operations. To all arc in the graph, various weights have been allocates which shows the combination of pheromone trail visibility and intensity. As well, throughput of such system has been taken as the determinant of pheromone trail visibility and intensity as the total processing time needed by the machining operations. As well ignore the local entrapment of the solution; few modifications as like: stagnation avoidance is made in the basic Ant Colony Optimization.